Insulin resistance contributes to the pathogenesis of several disease states including obesity and noninsulin-dependent diabetes mellitus (NIDDM). We have investigated the nature of the genetic factors that predipose to the development of insulin resistance. Mutations have been identified in the insulin receptor gene in patients with several genetic syndromes associated with insulin resistance. Some mutations cause insulin resistance by decreasing the number of receptors on the cell surface by one of several mechanisms: impairing receptor biosynthesis; impairing the transport of receptors to the plasma membrane; or accelerating the rate of receptor degradation. Other mutations cause insulin resistance by impairing one or more functions of the insulin receptor: decreasing the affinity of insulin binding; or impairing receptor tyrosine kinase activity. Recently, we have developed an animal model of this disease: a mouse in which the insulin receptor gene has been inactivated by homologous recombination. In order to investigate the significance of heterozygosity for a mutation in the insulin receptor gene, we have investigated the effect of mutations in the context of hybrid receptors formed by dimerization of a mutant half-receptor with a wild type half-receptor. In the case of the Leu-323 mutation (a mutation in the extracellular domain that inhibits insulin binding), the hybrid receptor bound insulin with normal affinity. This explains the observation that the Leu-323 mutation causes insulin resistance in a recessive fashion. In contrast, mutations in the intracellular domain (e.g., Ile-1153 mutation) generally inhibit receptor tyrosine kinase activity, and cause insulin resistance in a dominant fashion. However, Ile-1153 mutant receptors do not appear to dimerize with wild type receptors when both cDNAs are expressed in the same cells. The dominant effect of the Ile-1153 mutation may be explained by competition for binding of downstream effector molecules. While the prevalence of mutations in the insulin receptor gene is not known with certainty, most patients with NIDDM appear not to have mutations in that gene. Thus, studies have been initiated to screen for the presence of mutations in the gene encoding insulin receptor substrate-1 (IRS-1). Several variant sequences have been identified in the IRS-1 gene, and the prevalence of these variant sequences appears to be increased in patients with NIDDM. Studies are underway to determine whether these variant sequences impair the function of IRS-1.